© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    5. GPT2-Small
    7. Residual Similar CE E2E Recon
    9. 6-RES_SCEFR-AJT
    11. 5372
    Prev
    Next
    INDEX
    Explanations

    proper nouns and names of people or places

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-ajt
    Prompts (Dashboard)
    12,288 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    46,080
    Data Type
    torch.float32
    Hook Point
    blocks.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    apollo-research/Skylion007-openwebtext-tokenizer-gpt2
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    659
    +0.23
    0.9%
    547
    +0.14
    0.5%
    653
    +0.13
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.23
    0.19
    232
    +0.14
    0.13
    525
    +0.13
    0.11
    Negative Logits
    etheless
    -1.20
    ModLoader
    -1.15
    theless
    -0.91
    âĶĢâĶĢ
    -0.87
     awa
    -0.86
     Galileo
    -0.86
     Alibaba
    -0.83
    ãģ¯
    -0.83
    ?ãĢį
    -0.81
    éĹ
    -0.80
    POSITIVE LOGITS
    zen
    1.19
    stad
    1.14
    inski
    1.14
    ney
    1.11
    itz
    1.10
    ley
    1.10
    acci
    1.09
    owski
    1.09
    enberg
    1.08
    etti
    1.08
    Activations Density 1.493%

    No Known Activations